System Requirements for Cubit Estimating

Follow
Supported Windows Versions

Cubit Estimating is compatible with current versions of Microsoft Windows. Whether installing as a standalone application or a network client, consult your IT service provider to ensure a smooth setup. The standard installation is provided as an x64 (64-bit) application; an unsupported x86 (32-bit) version is available on request from help.buildsoft.com.au.

! Windows 10 — No Longer Supported: Microsoft ended support for Windows 10 on 14 October 2025. Cubit Estimating is no longer tested, certified, or supported on any edition of Windows 10. Customers must upgrade to Windows 11 Pro or newer to maintain compatibility, security, and product support.
Operating System Standalone Network Client Server Terminal / RDS Cloud (Windows)
Windows 11 Pro ✅ Supported ✅ Supported ❌ No ❌ No ⚠️ Limited
Windows 11 Home ✅ Supported ❌ No ❌ No ❌ No ⚠️ Limited
Windows 11 Pro Education ✅ Supported ✅ Supported ❌ No ❌ No ⚠️ Limited
Windows 10 Home / Pro ❌ No ❌ No ❌ No ❌ No ❌ No
Windows 8 / 7 ❌ No ❌ No ❌ No ❌ No ❌ No
macOS (Bootcamp / VM) ⚠️ Limited ⚠️ Limited ❌ No ❌ No ❌ No
Windows Server 2025 ❌ No ❌ No ✅ Supported ✅ Supported ⚠️ Limited
Windows Server 2022 ❌ No ❌ No ✅ Supported ✅ Supported ⚠️ Limited
Windows Server 2019 / 2016 ❌ No ❌ No ✅ Supported ✅ Supported ⚠️ Limited
Windows Server 2012 ❌ No ❌ No ⚠️ Limited ⚠️ Limited ❌ No
! Windows 10 is no longer supported by Microsoft or Buildsoft. Installations running on Windows 10 will not receive compatibility testing, updates, or support. Upgrade to Windows 11 Pro or newer is required.
i Cloud technologies involve varying environmental factors. Visit help.buildsoft.com.au for deployment guidance specific to your environment.
Hardware Recommendations

Buildsoft publishes three reference tiers based on workload intensity. Treat these as practical recommendations, not absolute pass/fail minimums.

🔴 Heavy Use
CPUIntel® Core™ i7-12700H or higher
RAM32 GB minimum
NetworkGigabit Ethernet+ or Wi-Fi 6
GPUNvidia GeForce RTX 3060 4 GB or better
Storage1 TB M.2 PCIe NVMe SSD
🟡 Medium Use
CPUIntel® Core™ i5-1135G7 or higher
RAM16 – 64 GB
NetworkGigabit Ethernet+ or Wi-Fi 4
GPUNvidia GeForce RTX 2050 4 GB or better
Storage512 GB M.2 PCIe NVMe SSD
🟢 Light Use
CPUIntel® Core™ i5-1135G7 or higher
RAM16 – 32 GB
NetworkGigabit Ethernet+ or Wi-Fi 3
GPUDedicated 4 GB video card or better
Storage512 GB M.2 PCIe NVMe SSD
¹ Intel or equivalent AMD processor  |  ² Nvidia or AMD equivalent video card
Graphics Requirements & Compatibility

Cubit Estimating requires a dedicated video card with OpenGL 2.0 support and up-to-date drivers. Note that a higher OpenGL version does not automatically guarantee compatibility — the card must specifically meet the OpenGL 2.0 requirement with functional driver support. Modern dedicated NVIDIA or AMD cards with at least 4 GB VRAM are recommended.

! OpenGL 2.0 is the specific requirement. A graphics card advertising a higher OpenGL version does not automatically guarantee compatibility with Cubit Estimating. The card must support OpenGL 2.0, have up-to-date drivers installed, and be correctly configured for the environment in use.
Graphics Compatibility Summary
Graphics Type Status Notes
Dedicated NVIDIA Graphics ✅ Supported GeForce RTX 2050 or newer recommended, 4 GB+ VRAM
Dedicated AMD Graphics ✅ Supported Equivalent Radeon cards, 4 GB+ VRAM, current drivers
Intel IRIS Integrated ⚠️ Limited Light use only. Reduced performance with large or complex projects
Snapdragon Adreno ⚠️ Limited Significant compatibility issues. Not recommended for production use
Legacy integrated (no OpenGL 2.0) ❌ No Unsupported. Hardware upgrade required
Processor Compatibility
✅ Fully Supported
  • Intel Core processors — 4th generation or newer
  • AMD Ryzen and equivalent processors
⚠️ Known Issues — Snapdragon ARM
  • Performance degradation and instability
  • Limited software compatibility
  • Graphics driver issues
  • Potential application crashes

Not recommended for production use of Cubit Estimating.

Storage Requirements
  • Minimum installation space: 2 GB of disk space
  • Free working space: At least 2.5× the current database size must be available as free disk space
  • Minimum headroom: Maintain at least 10% free disk space on the drive containing both the database and installation at all times
Display Requirements
  • Screen resolution: 1920 × 1080 (Full HD) or higher for both single and multi-screen setups
  • 4K single screen: Video card with 4 GB VRAM or more
  • 4K multi-screen: Video card with 8 GB VRAM or more
  • Scaling: 100% or higher supported, including 125%+ DPI scaling
Performance Impact Factors

The following conditions increase load on the CPU, RAM, storage, and graphics — and are the primary reason real-world environments often require a higher hardware tier than the baseline suggests:

Large plans and drawing files
Detailed PDFs and scanned images
Detailed drawing layers
Multiple open jobs simultaneously
Multiple users on the same job
Concurrent applications running on the same machine
A large number of reference items in a job
Hardware Benchmarking Tools

Tools for comparing your hardware against Buildsoft's published recommendations:

Overview

Running Cubit Estimating on a Terminal Server or Remote Desktop Server places higher demands on shared infrastructure than a standalone workstation. The specifications in this section are the recommended baseline for a stable, multi-user deployment. Requirements will vary based on project complexity, concurrent user count, and other applications sharing the environment.

! Recommended specifications may vary depending on the version of Cubit Estimating, project complexity, and concurrent user count. Consult Buildsoft Support in collaboration with your IT team for personalised specifications tailored to your environment.
i Virtual environments are supported for use with Cubit Estimating version 10 or above. Hyper-V and VMware are specifically supported virtualisation platforms.
Processor

Choose a high-performance multi-core processor capable of handling demanding calculations and rendering across multiple concurrent Cubit Estimating sessions.

  • Intel Xeon — server-grade multi-core processors
  • AMD EPYC / Ryzen — or equivalent in virtual infrastructure
  • High core count and strong single-thread performance are both important for multi-session workloads
vCPU Planning Guidance
  • Scale vCPU allocation with concurrent user count
  • Allow headroom for OS and background process overhead
  • Avoid over-provisioning vCPUs beyond the physical core count on the host

For processor compatibility on workstations (Intel Core, AMD Ryzen generations, Snapdragon), see the Workstation & Standalone tab.

Memory (RAM) & Server Sizing Calculator

Enter the number of concurrent Cubit Estimating users to calculate the minimum RAM and GPU VRAM required for Cubit Estimating alone. These figures cover Cubit Estimating only — your server will require additional RAM for the operating system, Windows Server RDS, and any other applications.

Cubit Estimating requires 2 GB RAM and 2 GB dedicated GPU VRAM per concurrent user.
Concurrent Cubit Estimating Users
concurrent users
Enter the number of concurrent users above to calculate RAM, VRAM, and server sizing.
Storage

Use fast, reliable storage for the OS and Cubit Estimating installation. Storage performance directly affects load times, file speeds, and overall responsiveness across all concurrent sessions.

  • Operating system drive: SSD — fast read/write is essential for a responsive server experience
  • Cubit Estimating installation: SSD recommended
  • Project / job files: Sufficient capacity for current and archived Cubit Estimating project files
⚠️ Backup & Free Space Requirements
  • Maintain at least 2.5× the total database size as free disk space for backup operations
  • Maintain an additional 10% of total storage as free disk space to accommodate growth

Insufficient free space can cause backup failures, database corruption, and application errors.

Graphics

Cubit Estimating relies on graphics processing for plan rendering, PDF display, and visualisation. In a terminal server environment the GPU is shared across all concurrent sessions — correct selection and configuration is critical.

! OpenGL 2.0 is the specific requirement. A higher OpenGL version does not automatically guarantee compatibility with Cubit Estimating. The card must specifically support OpenGL 2.0, have correct vendor drivers installed on the server, and be properly configured for the remoting stack in use.
  • Professional-grade GPU from NVIDIA or AMD
  • Dedicated VRAM — 2 GB per concurrent user where possible. A GPU with 10 GB VRAM supports up to 5 concurrent users; 20 GB supports up to 10 users
  • OpenGL 2.0 support with up-to-date vendor drivers installed and active on the server
  • GPU must be correctly configured for the display/remoting stack in use (RDS, Citrix, VMware Horizon, etc.)
⚠️ Configuration Notes
  • Windows Server does not include OpenGL support by default — it is provided by the GPU driver
  • Remote Desktop Services may disable hardware graphics acceleration unless a supported GPU and correct driver are present
  • Compute-only GPU deployments (no WDDM display driver) are not suitable for Cubit Estimating
  • Verify OpenGL 2.0 is available and functional after driver installation before deploying to users
Server GPU Suitability
Graphics Type Suitability Notes
Professional NVIDIA (Quadro / RTX Ada / A-series) ✅ Recommended Dedicated VRAM, OpenGL 2.0+, designed for multi-session server use
Professional AMD (Radeon Pro / FirePro) ✅ Recommended Dedicated VRAM, OpenGL 2.0+, with current AMD drivers
NVIDIA A10 with vGPU / RTX Virtual Workstation ✅ Suitable Designed for graphics-rich VDI. Requires vGPU-capable WDDM graphics stack
Consumer NVIDIA / AMD (GeForce / Radeon) ⚠️ Limited May function on small deployments — not designed or tested for server multi-session use
Integrated / Intel IRIS ❌ Not suitable Insufficient for multi-user terminal server deployments
Compute-only GPU (no WDDM / display driver) ❌ Not suitable Graphics APIs including OpenGL are unavailable without a WDDM-based graphics stack
VRAM Planning Reference (2 GB dedicated per concurrent user)
1 user
2 GB
2 users
4 GB
3 users
6 GB
4 users
8 GB
5 users
10 GB
6 users
12 GB
7 users
14 GB
8 users
16 GB
9 users
18 GB
10 users
20 GB
Allocate 2 GB of dedicated VRAM per concurrent Cubit Estimating user where possible. Add overhead for the display driver and OS frame buffer on top of user session totals. Where VRAM is constrained, prioritise fewer users per GPU rather than over-provisioning sessions.
Server & Workstation GPUs — OpenGL 2.0 Compatibility
Based on vendor specifications only — not tested by Buildsoft. The cards listed below are documented by their manufacturers as supporting OpenGL 2.0. Compatibility with Cubit Estimating has not been independently verified. Use in production environments is at the customer's discretion.
i Windows Server and OpenGL: Windows Server does not include OpenGL support by default — it is provided by the GPU driver. Three conditions must be met: (1) a GPU that supports OpenGL 2.0, (2) the correct vendor driver installed, and (3) remote desktop acceleration enabled where applicable.

Common in rack workstations and blade servers.

  • Quadro FX 500 / 700 / 1000
  • Quadro FX 1400 ⭐
  • Quadro FX 1500 / 2000 / 3000
  • Quadro FX 3450 ⭐
  • Quadro FX 3500 ⭐
  • Quadro FX 4500 / 4500 X2 ⭐

⭐ Recommended for PCIe compute nodes, rack workstations, and blade server deployments.

Verify OpenGL mode is enabled — some Tesla configs disable display output by default.

  • Tesla C870
  • Tesla S870
  • Tesla C1060 ⭐

⭐ Recommended. Used in compute nodes and older server infrastructure. Confirm OpenGL 2.0 is enabled via driver before deployment.

OpenGL 2.0-era cards with programmable shader support.

  • Radeon 9500 / 9550 / 9600 / 9700 / 9800
  • Radeon X300 / X600 / X700
  • Radeon X800 / X850

Compatible where a PCIe slot is available and drivers are current.

Network

A stable and high-speed network connection is critical for smooth collaboration and data transfer in a terminal server environment.

  • Use a reliable Ethernet connection with adequate bandwidth to handle the expected user load
  • Keep the database and Terminal Server on the same local network — avoid remote or WAN links between the database server and the terminal server host
  • Network latency and packet loss directly impact the responsiveness of Cubit Estimating sessions for remote users
! Placing the Cubit Estimating database on a remote server connected via a WAN or slow link is a common cause of performance problems and file integrity issues.
Operating System

Use a server-grade operating system designed for terminal server environments to ensure stability, security, and efficient resource utilisation.

Operating System Terminal / RDS Status Notes
Windows Server 2025 ✅ Supported Recommended for new deployments
Windows Server 2022 ✅ Supported Recommended for new deployments
Windows Server 2019 / 2016 ✅ Supported
Windows Server 2012 ⚠️ Limited Limited support only. Plan migration to a current Server version
i Windows 11 Pro and Pro Education support Cubit Estimating as standalone or network client installations — they are not suitable as terminal server hosts. Use Windows Server editions for RDS / terminal server deployments.
Virtualisation

Virtual environments are supported for Cubit Estimating version 10 or above. Virtualisation allows better resource management, scalability, and easier maintenance of the terminal server environment.

  • Microsoft Hyper-V
  • VMware vSphere / ESXi
  • Other hypervisors supporting Windows Server guest VMs with GPU passthrough or vGPU

Cubit Estimating version 10 or above is required for virtual environment support.

⚠️ GPU in Virtual Environments
  • The virtual machine must have access to a GPU with OpenGL 2.0 support and up-to-date drivers
  • GPU passthrough or vGPU (e.g., NVIDIA GRID / vGPU, AMD MxGPU) must be correctly configured
  • Validate that OpenGL 2.0 is functional within the guest OS after driver installation — a higher OpenGL version in the driver does not automatically guarantee compatibility with Cubit Estimating
Azure VM Guidance

Azure Virtual Desktop supports GPU acceleration for application rendering and remote frame encoding. Session hosts must use supported GPU-optimised Azure VM sizes with the correct Azure-distributed graphics drivers.

! For NVIDIA-backed Azure VM sizes, NVIDIA GRID drivers are required for GPU acceleration. CUDA drivers do not provide the required Windows UI or application acceleration in these scenarios.
! NVv4 retirement: Microsoft is retiring NVv4 VMs on 30 September 2026. Avoid NVv4 for new deployments and migrate existing workloads to NVads V710 v5 or an equivalent alternative.
Recommended Azure VM Families
Azure VM Family Use Case Status Notes
NVadsA10 v5 Best overall choice for graphics-enabled Cubit Estimating sessions ✅ Recommended NVIDIA A10 based, suited to graphics-rich virtual desktops
NVads V710 v5 AMD-backed option for right-sized sessions ✅ Recommended 1/6 GPU to full GPU options, 4–24 GiB frame buffer
NVv4 Legacy only ⚠️ Retiring Sep 2026 Avoid for new deployments. Migrate to NVads V710 v5
Suggested Azure Sizing
Use Case Suggested Starting Point
1–3 heavier users NVadsA10 v5 fractional GPU profile
3–8 mixed users NVadsA10 v5 or NVads V710 v5 mid-tier profile
8–15 light to medium users Multiple GPU-enabled session hosts rather than a single large host
Larger environments Scale out horizontally and validate with pilot users
Azure cloud remains ⚠️ Limited support per Buildsoft's published position on Windows-based cloud services. Validate your deployment with Buildsoft Support before go-live.
Citrix & Virtual Desktop Infrastructure (VDI)

Cubit Estimating should be treated as a graphics-accelerated Windows desktop workload — not a compute-only workload — when planning any VDI or virtual desktop deployment. Validate against your intended session density and drawing complexity before go-live.

GPU-enabled delivery can be implemented using GPU passthrough or virtual GPU (vGPU).

Passthrough support

  • Citrix Hypervisor
  • Nutanix AHV
  • VMware vSphere / ESX
  • Microsoft Hyper-V (Discrete Device Assignment)

vGPU support

  • Citrix Hypervisor
  • Nutanix AHV
  • VMware vSphere
  • NVIDIA GRID / vGPU and AMD MxGPU-backed acceleration

Use vGPU for shared density and easier scaling. Use passthrough for maximum consistency with a small number of users.

VMware Horizon

  • GPU passthrough or vGPU via VMware vSphere / ESXi
  • NVIDIA GRID and AMD MxGPU supported for virtualised graphics acceleration
  • Ensure OpenGL 2.0 is available within the guest VM via the configured driver stack

Azure Virtual Desktop

  • GPU acceleration supported for rendering and remote frame encoding
  • Only Azure-distributed graphics drivers are supported on session hosts
  • NVIDIA GRID drivers required for NVIDIA-backed N-series VMs — CUDA drivers are not sufficient

Use GPU-enabled NV-family session hosts with Azure-distributed graphics drivers for Azure Virtual Desktop deployments.

i All VDI and virtual desktop deployments should be validated with a pilot before full rollout. Session density, plan complexity, and remoting protocol overhead all affect real-world GPU requirements beyond the planning estimates.
Security & Backup

Cubit Estimating relies on the security practices of its host environment. Ensure the following are in place:

  • Appropriate security perimeters and access controls for the terminal server environment
  • User access settings and permissions scoped to Cubit Estimating data and application directories
  • Active monitoring of user activity and resource utilisation
  • Establish a comprehensive backup strategy protecting critical Cubit Estimating project files
  • Implement redundancy for critical components — power supplies and network connections
  • Maintain at least 2.5× the total database size as free disk space for backups
  • Maintain an additional 10% of total storage as free space for growth and changes
Monitoring & Management

Use appropriate tools to monitor system performance, user activity, and resource utilisation so that issues can be identified and resolved before they impact Cubit Estimating users.

System Performance
  • Monitor CPU, RAM, and storage utilisation across all concurrent sessions
  • Watch for resource contention under peak user load
User Activity
  • Track concurrent session counts against hardware capacity
  • Alert on abnormal session behaviour or resource spikes
Storage & Database
  • Alert before free disk space falls below the 10% or 2.5× database thresholds
  • Verify backup jobs complete successfully after each run
Backup Storage Calculator

Use this calculator to determine the minimum free disk space required for your Cubit Estimating database backups. Enter your current database size and the calculator will apply Buildsoft's published storage requirements.

i Buildsoft requires at least 2.5× the total database size as free disk space for backup operations, plus an additional 10% of total drive capacity as headroom for growth.
Enter Database Details
gigabytes (GB)
gigabytes (GB)
Enter your database size above to calculate storage requirements.
Buildsoft Storage Rules Reference
Maintain at least 2.5× the total database size as free disk space on the drive containing the database. This ensures sufficient room for backup files and temporary working space during the backup process.
Maintain a minimum of 10% free disk space on the total drive / volume at all times. This headroom accommodates growth in project files, OS operations, and changes in storage requirements.
Allow a minimum of 2 GB for the Cubit Estimating installation itself, separate from the database and backup space calculations above.

Questions about your environment or deployment? Contact the Buildsoft support team at help.buildsoft.com.au

3 out of 3 found this helpful

Comments

2 comments
  • Awesome article Michael. Very helpful. Thank you.

    0
    Comment actions Permalink
  • Hi Michael,
    How much space would be required for network client storage? We have the Enterprise version of cubit and have been using it since early 2022 and are currently using 672GB. We just increased it from 500GB to 800GB last month and it just keeps on growing. Thanks.

    0
    Comment actions Permalink

Please sign in to leave a comment.